Reviewer

Overview
Non-reproducible
Methodology · Analyses · References · Findings · Conclusions
Created Mar 13, 2014 · Edited Nov 21, 2014
Review Contributors: Tommy Lo, Florence Tang, Kenneth Ka-Ho Lee
Methodology
No
Are the tools and methods appropriately applied, and are the data correct?
The methods for generating STAP cells and their maintenance as reported in the paper are very novel. However, the readers would find it difficult to validate some of the findings.

For instant, there are some confusing steps in the procedures and the same confusion was also observed in their recent follow-up publication in “Protocol Exchange” dated 5th March 2014.

We have attempted to establish whether it was possible to create STAP cells using Haruko Obokata et al’s protocols. We first obtained Pou5f1-GFP mice and proved that they were capable of expressing Oct4-GFP by examining their germ cells inside their gonads. All of the mice that we have used contained germ cells that were Oct4-GFP+ - as determined by confocal microscopy and Flow cytometry (Figure 1A-C). We used freshly isolated 6-week old (adult) splenocytes and treated them with pH5.7 HBSS for 25min at 37oC as cited in Obokata’s paper. These cells were then maintained in culture for up to 7 days. During the whole culture period, we did not observe any Oct4-GFP expression. The results were validated by qPCR where we confirmed no induction in Oct4 expression but instead some suppression of Sox2 and Nanog expression (Figure 1D-F). We also found similar results for primary fibroblasts isolated from lung explants of 1 week (neonate) and 6 week (adult)-old Pou5f1-GFP mice (Figure 2). We also attempted the experiments in human umbilical cord perivascular (HUCPV) progenitor cells and again it did not work. We did not see Oct4 expression (Figure 3). It was while we were conducting these experiments that a more detailed protocol was published in Protocol Exchange, so we decided to do additional sets of experiments. CD45+ splenic lymphocytes were isolated from the spleen of one-week-old Pou5f1-GFP mice. These cells were treated with HBSS, adjusted to pH 5.7 with HCl, for 25min at 37°C (according to their newly released protocol). We cultured these cells for 6 days and did not see any Oct4-GFP expression. This was validated by qPCR (Figure 4). The experiment was repeated twice.
In conclusion, we have not been able to produce “STAP” cells using the protocols reported in Obokata’s publications.
Analyses
Yes
Have the analyses, including any statistical analyses, been performed appropriately and rigorously?
Their analyses were comprehensive and easy to follow. Also a large number of repeated experiments was performed and reported (n>20 for some experiments) - make the results plausible and convincing.

However, we have not been able to replicate some of them.
References
Yes
Do the authors refer to the latest research in the field, and is it correctly referenced?
In the context of the study, the references were relevant and meaningful.
Findings
Yes
Do the authors provide an interesting discussion, and are the findings novel?
The authors highlighted that various stress-inducing factors1, like hypoxia, have been reported to up-regulate Oct4 expression in various cell types - which support their findings. However, they mentioned that the potential of establishing methods to generate iPSCs using these factors has still not yet been exploited. Consequently, this stimulates great interest and awareness in readers to discovery other novel and simple methodologies for cell reprogramming.

1. Mathieu J, Zhang Z, Nelson A, Lamba DA, Reh TA, Ware C, Ruohola-Baker H. (2013) Hypoxia induces re-entry of committed cells into pluripotency. Stem Cells 31(9), 1737-48
Conclusions
Partly
Are the interpretations and conclusions justified by the results?
The results and analyses support the authors’ claims in this publication. However, the ease and simplicity of their method for generating STAP cells from various stressors and cell types has left the readers in doubt.

We have tried our very best to generate STAP cells using their protocol and it appears that it is not as simply and reproducible as we expected. So whether the techniques really work, still remains an open question?
Supporting resources
Source
Source
Source
Source

Comments (61)

  • Mention of this review is featured on BBC News http://www.bbc.co.uk/news/health-26576368
  • Sergey Kiselev · Vavilov Institute of General Genetics
    Dear Kenneth i beg my pardon. I did not study your past pictures. You are absolutely right. Sorry again.
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    Dear Walton.  Excellent suggestions.  It would be great if  Dr Haruko Obokata could come to my lab and show me and my students how STAP should be properly done.  Perfectly possible that we missed some critical steps.  All reagents and mice are ready for her or any of her lab members to come to Hong Kong.  Really want STAP to work. My very best wishes to her during this difficult time.

    Dear Sergy.  No problem. We all miss fine details.
  • Jeanne Pawitan · University of Indonesia
    Dear Prof. Lee,
    If Dr. Obokata would like to come to your lab, I suggest that you ask her to bring her mice along with her. I am almost sure that you will be successful in your attempt if you use her mice and acid bath. I wish you good luck.

    However, as Mr. Tommy Lo has said, the whole project will be meaningless.
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    Dear Jeanne,

    We use  CBA-Tg(Pou5f1-EGFP)2Mnn/j transgenic mice from The Jackson Laboratory, which is basically what most people use.  Unless Obokata's group developed their own transgenic line which is very labour intensive and not worth the expense!

  • Jeanne Pawitan · University of Indonesia
    Dear Prof. Lee,
    I learned from Technical tips for STAP cells in Nature Protocol Exchange that Obokata et al used an Oct-3/4-EGFP transgenic mouse line from Ohbo et al (Dev Biol 2003), I supposed that they did not develop the mouse line themselves, but Ohbo et al did.. 

    The gene construct might be different from that in your transgenic mice, what's your opinion?


  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    Dear Jeanne,

    The type of transgenic mice used is important but, at the end of the day, the results should be further validated by qPCR or immunohistological staining- which we did.
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    New STAP Protocol released.  Been asked by Alexey Bersenev to try it out.

    • Source
      [Show abstract] [Hide abstract]
      ABSTRACT: We have found this to be an effective protocol for generating STAP cells in our lab, regardless of the cell type being studied. The refined protocol below takes elements from the two most effective approaches described in our Jan 31, 2014 article published in Nature (Obokata et. Al., Stimulus triggered fate conversion of somatic cells into pluripotency. Nature 505. 641-647, 2014). We have found this refinement to be useful as a generic protocol to generate STAP cells from a “generic” source of cells. It is very important that each step be performed precisely as described. The protocol is extremely simple, but will vary slightly, if you are starting with tissue rather than a cell suspension. It also will vary depending upon the cell type or tissue with which you are starting. It is important to not skip any steps. It is especially important to triturate the cell suspension for a minimum of 30 minutes, until the suspension can be easily triturated up and down the reduced bore pipettes of the smallest orifices. We first describe the protocol when starting with a suspension of cells, and then describe additional steps necessary when starting with a soft tissue.
  • Jeanne Pawitan · University of Indonesia
    Dear Prof. Lee,
    Are you going to try it out? If so, I wish you good luck.
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    Yip, We will!
    And maybe we get another invitation by Researchgate to  review the protocol released.
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    Dear Professor Lee

    Thank you for submitting your comment on one of our published papers to the Brief Communications Arising section. Regretfully, we cannot offer to publish it.

    This section of Nature is extremely oversubscribed, so we can consider only a very few of the critical comments we receive. Our main criterion for consideration is the degree to which the comment challenges the main conclusions of the published paper in question.

    In the present case, while we appreciate the interest of your comments to the community, we do not feel that at this stage they challenge key data or conclusions of the paper by Obokata et al., and therefore we cannot offer to consider your paper for publication in our Brief Communications Arising section.

    Although we cannot offer to publish the submission as a Brief Communication Arising, you may wish to use our online commenting facility (see http://www.nature.com/nature/journal/v464/n7288/full/464466a.html).
    To post a comment, scroll to the bottom of the online html version of the article you want to comment on. When using the online commenting facility for the first time, you will need to agree to the terms and conditions before a comment can be posted. Using this option would retain the linkage of your comment to the original paper, and would allow for further discussion by the community on the points you have raised.

    I am sorry we cannot be more positive on this occasion.

    Yours sincerely

    Francesca Cesari, Ph.D.
    Senior Editor
    Nature
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    This is what happens when you go peer review!  How come my data  "we do not feel that at at this stage they challenge key data or conclusions of the paper by Obokata et al".

    If the first step of making STAP can  not be replicated then both papers are fundamentally flawed!!

    This is Nature not admitting their review process have problems.



  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    Read my interview on Wired about Peer review verses  scientific social network.

    http://www.wired.co.uk/news/archive/2014-03/14/research-gate-kenneth-stem-cell-debunk

  • Jeanne Pawitan · University of Indonesia
    Dear Prof. Lee,
    Indeed, Nature peer review process is full with problems. Not only the result can not be replicated (at least until now), there are many other serious problems, which can be seen in the website:

    Moreover, the investigation has lead DR. Obokata to try to withdraw her dissertation:



  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    Attempts to replicate STAP using newly released protocol by Charle Vacanti's group:

    We are open-minded and hopefully it will work this time.  I have set up a team of 4 to do the experiments.  I will personally make the glass pipette with a 150um ED and ID of 75um.  I have forged similar pipette for cloning mice while at Edinburgh University for many years.

    Let's see how it goes.  I will be posting regular update on this page.  .


    • JPG
      IMG_2694.JPG

    Modified Mar 24, 2014 by the commenter.

  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    24th March 2014  7:30pm Hong Kong time.

    We have sufficient lung fibroblasts from 5 day old postnatal Pou5f1-GFP mice for STAPing.  These cells are passage 2.  Tomorrow, we will make the pipettes for robust titation and perform the "acid bath" processess.
    • jpg
      Lung fibroblast.jpg

    Modified Mar 30, 2014 by the commenter.

  • Maria Vicenta Camarasa · Fundación Caubet-Cimera Centro Internacional de Medicina Respiratoria Avanzada
    I will read the original paper carefully, I have requested full copy here.... nevertheless, my opinion here is that if that pulse is robust enough with x factors in the culture system, the result will be reproduced in any lab after several attempts... obviously official journals won't pubblish all negative data nor comments, awaiting for more positive data to appear... if that doesn't happen, the thread will be forgotten and other roads pursued... obviously medical field needs to obtain pluripotent cells reproducibly without genetic modification for the whole field to advance and settle down.... hopefully electrical pulses, or just protein or RNA addition will work... those areas had initial papers but not establishment in the research community... I can understand how these protocols are prone to being not reproducible... 'catching pluripotency' has to be intrinsically difficult, knowing that in nature is so transient... but we need to continue figuring out how to.... just keep trying, look at the pictures not at the cells directly to study them meanwhile cells grow... and select select select... then be consistent on methods, and result will appear... you probably won't be asked to tell how many attempts were unsuccessful, just the body of experiments giving a reasonable mean of desired results... that's science, I guess....
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    25th March 2014  10:30pm Hong Kong time.

    Today, we made the glass pipettes for titrating the fibroblasts before acid treatment. Two sets of pipettes were made; one with 150µm and other set with 75 µm.  We also performed pilot studies with the pipettes – since the bore size is very small.  We tried using a Pipetman and a mouth pieceto determine which is the better method.  It was decided a 5ml Pipetman was the better choice because it was exhausting trying to titrate by mouth.

    • pdf
      Pipettes.pdf

    Modified Mar 30, 2014 by the commenter.

  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    26th March 2014  5:30pm Hong Kong time.

    We have attempted to generate STAP cells using Vacanti's newly released protocol.

    What we did today:

     Lung fibroblasts (passage 2) were trypsinized, pelleted and washed with HBSS.

    ~0.5x106 cells were then resuspended in “Sphere Media” (DMEM/F-12, 1%PS, 1xB27, 1,000U/mL LIF). -ve control.

    6x106 cells were triturated. Briefly, the cells were suspended in 2mL of HBSS (cell conc. = ~0.8x106cells/mL). Using glass pipettes, pre-coated with chilled HBSS, with different lumen sizes for trituration: 5 minutes with standard 9” glass pipette. 10 minutes through larger lumen (124μm). Then 15 minutes through small lumen (61μm). During the process, no air bubbles were allowed to form.

    After that, additional HBSS was added drop-wise and air bubbles avoided, and made to final volume of 20mL. The cells were pelleted by centrifugation at 1,200 rpm for 5 minutes.

    ~0.5x106 cells were then resuspended in “Sphere Media” (DMEM/F-12, 1%PS, 1xB27, 1,000U/mL LIF). “Trituration only and no acid treatment” –ve Control

    ~1.1x106 cells were then re-suspended in HBSS at pH 5.4 at room temperature, at concentration of 2x106 cells/mL. The pH rise to around pH5.8. The pH of the cell suspension was then titrated to pH5.67 with diluted HCl in HBSS, immediately before sealing a conical tube for incubation.

    The acid bath and non-acid bath cell suspensions were incubated at 37°C for 25 minutes, followed by centrifugation at 1,200rpm for 5 minutes.

    Finally, cells were re-suspended in “sphere media” and seeded on non-adhesive culture dishes.

    Note:

    1. Non-adhesive 35-mm culture dishes were used. (SPL, cat#:11035)
    2. Seeding density for producing the STAP cells was ~1.2x105 cells/cm2
    3. No EGF, bFGF or heparin was added in the ‘Sphere Media’ à Obotaka’s original recipe should be able to support and maintain cell reprogramming

    • pdf
      Presentation1.pdf

    Modified Mar 30, 2014 by the commenter.

  • Elliott Batte · University of Colorado
    Dr. Lee,

    Thank you for posting all of these details. I know many people are following intently. 

    Elliott
  • Srav Gopa · Keio University
    '6x106 cells were trituated. Briefly, the cells were suspended in 2mL of HBSS (cell conc. = ~0.8x106cells/mL). Using glass pipettes, pre-coated with chilled HBSS, with different lumen sizes for trituration: 5 minutes with standard 9” glass pipette. 10 minutes through larger lumen (124um). Then 15 minutes through small lumen (61um). During the process, no air bubbles were allowed to formed.'

    Typo? - 1.6 x 106 cells to start with, right?

  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    Hi Sra,

    Unfortunately this site does not accommodate superscript or Greek alphabets. 

    It should be 0.8 X 10 to the power 6 or  0.8 million cells.   Thank you for pointing this out. 

    The same  also apply for ~1.2x105 cells/cm2 which should be 1.2 X 10 to the power 5 and the 2 after cm

    Modified Mar 27, 2014 by the commenter.

  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    27th March 2014  11:00pm Hong Kong

     Dear All,

    It is now 1day after our transgenic (Oct4-GFP) fibroblasts have been mechanically titrated and acid bathed.  Below you will find the results of our confocal analysis.  Enjoy!


    • pdf
      STAP_Vacanti_Protocol.pdf

    Modified Mar 30, 2014 by the commenter.

  • Jeanne Pawitan · University of Indonesia
    Dear Prof. Lee,
    Are you sure that they are auto-fluorescence from necrotic cells?
    I really hope that they represent Oct-GFP expressing cells.


  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    Dear Jeanne,

    Yip, I am very sure because the cell indicated by the green arrow in Figure C and C’ has a slight green grow but the DAPI staining revealed the nucleus is not intact i.e. a dying or dead cells.

    Anyway, time will tell as we expect more and more necrotic cells will form during culture. The third day of culture is the critical period as reported by Vacanti.

    We will validate all our results and conclusions by qPCR.

    Modified Mar 28, 2014 by the commenter.

  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    28th March 2014  8:00pm Hong Kong

    Dear All,

    It is now 2 days after our Oct4-GFP fibroblasts have been mechanically titrated and acid bathed.  Below you will find the results of our confocal analysis.  Enjoy!

    Our manipulation and observations today:

    Trituration was performed twice, same as day 1.  0.5mL of ‘Sphere Media’ was added to each of the 35mm dishes, as described in Vacanti Lab’s protocol.

    Although no statistics was cited by Vacanti, the cell density decreased drastically. We estimated that there was a 50% decrease in cell number. In the original paper reported in Nature, such decrease in cell count was reported for day 2, which is inline with our current experiment.  Day 3 will be critical as this was the time Oct4-GFP expression was reported for STAP cells.  If we find that the cell number decreased even more drastically in our cultures, we will harvest some of the cultures and use them directly for qPCR analysis.



    • pptx
      STAP day2.pptx

    Modified Mar 30, 2014 by the commenter.

  • Jeanne F Loring · The Scripps Research Institute
    Ken:
    I see in your earlier photos that the green fluorescence is not associated with DAPI, so it is likely to be autofluoresence associated with dying or dead cells.  Is the fluorescence emission limited to the GFP emission spectrum or is it the broader sort typical of autofluoresence?
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    30th March 2014  8:00pm Hong Kong (been so busy even got the year wrong in all previous posts!! Really embarrassing).

    Weekend, but we still managed to get access to the confocal microscope. Below you will see the results of the 3 day-old Control and STAP cultures using Vacanti’s Protocol.  Enjoy!



    • pdf
      STAP_vancanti_day3.pdf
  • Paul Knoepfler · University of California, Davis
    Thanks, Ken. For reporting your studies here. This is extremely helpful.
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

     

    Dear All,

    I am shocked and amazed by the qPCR results for the 3 day-old  control and STAP cultures.  Totally speechless!

    ENJOY!

    The STAP Saga continues...........


    • pdf
      STAP_VacantI_PCR.pdf
  • Sergey Kiselev · Vavilov Institute of General Genetics
    Great joke for April Fools!
    http://www.japantimes.co.jp/news/2014/04/01/national/high-profile-stem-cell-papers-contain-fraudulen...

    Go on, Ken, please...
  • Jeanne Pawitan · University of Indonesia
    Dear  Prof Lee,
    It seems that trituration only may yield STAP cells??

    I hope that you go on culturing the STAP cells in ACTH or FGF4 containing medium, to make them into STAP stem cells. 

    Good luck
  • Private Profile
    What a great news. Well-done!
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer
    Dear All,

    This not an April Fool trick!.  The negative control came up "positive" while the acid bath came up "negative".  Looks like mechanical trituration could induce STAP cells!?!?!?  However, this the first time we did this experiment and will require several more rounds to validate.


  • Walton Malcolm Byrnes · Howard University
    Now the race is on! This is very exciting, Dr. Lee.
  • Julien Maruotti · Phenocell
    Dear Dr. Lee,

    Sometimes stressed cells can express a wide range of genes, not necessarily connected to a given phenotype. Although it is interesting to note that SOX2 is not up-regulated in the trituration only sample, contrary to OCT4 and NANOG, maybe other markers -not necessarily related to pluripotency- would show similar profile of up-regulation. Do you plan to also check additional genes?
    Anyway it looks very intriguing.

  • Paul Knoepfler · University of California, Davis
    Let's see how this develops, but I remain skeptical that this is a specific induced pluripotency-related event related to trituration and that what you are seeing here is STAP cells. I hope I'm wrong and it is something real on the STAP front, but I doubt it. Thanks again, Ken, for all the hard work that your lab is doing!
    Paul
  • Andrés M Bratt-Leal · Parkinson's Association of San Diego, The Scripps Research Institute
    Hi Dr. Lee,

    Dividing by a number close to zero can lead to large increases in fold change even without significant gene expression. Can you post a comparison to pluripotent stem cells or the fold change compared to your housekeeping gene in the untreated and treated samples?

    Thanks,
    Andres
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    Dear All,

    Mechanical trituration is not hard to do.  Can some other labs please try and see whether using Vancanti's protocol of just mechanical trituration could induce expression of the stemness markers that I just mentioned? 

    Potentially, expression of these pluripotent markers could be the bi-product of un-regulated gene expression by the dying or stressed cells. I agree 100% with Paul Knoepfler's comments.

    I am not claiming that "STAP" cells exsist  - only presenting the results of our research as it is - which is open to interpretation.  Please, don't Hype up this data!

    Modified Apr 2, 2014 by the commenter.

  • Henry Chang · National Institutes of Health
    I'm not in the lab any more, but perhaps the acid treatment just pulls histones or repressors off DNA, that allows cells to become undifferentiated. Thus, I wonder if anyone has tried progressively lower pHs in media compatible with cell replication, which also results in chromatin disruption. Several days of culture would be required. To me, the stress hypothesis may be too narrow, but the baby should not be thrown out with the bath water!
  • Jichang Wang · Max-Delbrück-Centrum für Molekulare Medizin
    I do not think the qPCR data can tell anything.
    1) There are just about 10 folds in treated cells compared with the untreated cells, meaning just 3 CT-value difference between treated and untreated cell. I guess these differences just came from technical errors (RT and amplication efficiency) or unspecific PCR products (e.g. primer dimers).
    Dr. Lee, did you run a gel to check the qPCR products? Could you upload the raw qPCR data (CT values), please? Usually, CT value >= 37 means no amplification.
    2) The GFP signal seems to be autofluorescence coming from dead or dying cells. More precisely, western blotting is required for confirmation of expression of Oct4/Sox2/Nanog in treated cells.

    Modified Apr 2, 2014 by the commenter.

  • Srav Gopa · Keio University
    Dr. Lee - The figure from Day 3 (A and B) indicate that only dying (PI positive) cells show faint OCT-4GFP fluorescence. And since you started with 0.5 x 10 6 cells for the -ve control - most of them look unhealthy and in the process of dying. DO you know how good the quality of mRNA you obtained from these dying cells? qPCR typically amplifies a region of 100 bp!
    Also, may be  fluorescence may not be a reliable indicator (with auto fluorescence etc). As pointed out by Jichang Wang above , WB for the expression of Oct-4 and Nanog might be more reliable...
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    Thank you all for your excellent suggestions on improving the data.  Personally, I don’t think STAP cells exist and it will be a waste of manpower and research funding to carry on with this experiment any further. 

    I think this live-blogging is a good and bad idea:

    Good in the sense that it will stimulate interest in the scientific community, so helping to draw younger scientists into the stem cell field.

    Bad in the sense that you don’t have the time to think and carefully assess the data properly -   as everyone expect you to post the results as soon as it is generated. E.g. In my case a 10 folds increase in Oct4 and Nanog expression is definitely not sufficient and we need to see at least 100 folds as in my previous post (see above Figure3 for iPSCs). And for Sox2 it is only at 2 fold increase - which is even worst.

    I will no longer blog on this page any more.  I want to get back to doing my own science interest.

    Anyone want to collaborate with me to look at the function of the BRE gene which is a component of the BRCA1 and BRISC complex?

     

    Modified Apr 3, 2014 by the commenter.

  • Jeanne F Loring · The Scripps Research Institute
    Ken- Thank you so much for doing this- you have many fans!  It made all the difference to me to follow your experiments and results, and I felt like I was in the lab next door to you, hearing about your work every day.  I think this experience needs to be written about---in a journal.  Consider gathering some others who have tried this and writing a perspective article- for Science, perhaps- that describes the process of trying to reproduce a high impact result.  Meanwhile, look at the two short papers I've attached, sent to me recently by a friend, as a example of researchers working together to see if a key result was reproducible.
    Jeanne

    • pdf
      Tuszynski et al Cell response.pdf
    • pdf
      Steward et al-Cell 2014.pdf
  • Fulvio Celsi · IRCCS Ospedale Infantile Burlo Garofolo
    I really have to thank profoundly Prof. Lee for all the work and effort put here. It was beautiful to follow experiments and to be able to understand what the process would be going on.. I agree with  Jeanne just before me...it is worth to put it down in a paper. 
    Thanks again Prof. Lee!
  • Kenneth Ka-Ho Lee · The Chinese University of Hong Kong
    Reviewer

    Thank you all for your kind comments.  I said I was not going to live blog anymore but I also hate to leave the STAP story incomplete.

    Previously (1st of April), we have demonstrated that pipetting Oct4-GFP lung fibroblasts through different size glass pipettes for 3 days can elicit Oct-4 and Nanog expression by approximately 10 folds and Sox2 by 2 folds. The expression level was not high enough to label them as pluripotent stem cells.

    We have now completed a second set of experiments.  We did not do any acid-bathing this time - only mechanical pipetting.  We pipetted the cells for 30 mins and then grew them on non-adhesive culture plates.  On successive days, the cells were pipetted twice (5 mins) a day.  The same culture condition as in the first set of experiment.  We established by qPCR that this hash treatment induced Oct-4 expression by 2.5 folds.  It did not induce Sox2 and Nanog expression (Table 1).

    Since there were still cells alive at the end of 3 days, we decided not to do only more pipetting and just let these cells  grow on the non-adhesive culture plates for a further 3 days.   At the end of 6 days of culture, the cells were harvested for qPCR.   We found that Oct-4 expression was maintained by approximately 2.5 folds.  Like in day 3, Sox2 and Nanog expressions were not induced (Table 2).   

    #### Hypothesis: Perhaps treating these cells with Valporic acid to open up the histones might facilitate a better outcome in terms of expression of the stemness markers?

    There is a Chinese saying that:  when we are about to die, we can release a final burst of energy that allow us to finish off some of the things that we want to do before we leave this world “迴光返照”.  Could it be that the fibroblasts that we mechanical pipetted to the point of death “reawakens” and start randomly express some of the stem cell markers?  These cells are definity not STAP cells.  Perhaps I should call them “April fool” cells or even “reawakening” cells (Joke!).

    Even better, I could call them “Easter” cells since Jesus was reawakened from death and the Easter holidays are fast approaching us (Joke!)?

    I am going to take a page from Paul Knoepfler's Blog and ask you to vote.

    What should I call the cells that I created here?

    1. STAP cells

    2. April Fool Cells

    3. Reawakening cells

    4. Easter cells

     

    • pdf
      Experiment 2_vacanti.pdf
  • Takashi J Ozaki · Recruit Communications Co.,Ltd.
    Hi Prof. Lee,


    Thank you very much for the absolutely interesting result, and also for friendly communication with me on Twitter, although I'm just a lay observer from the other field.

    Indeed, some researchers and/or experts in Japan also suggest (on Twitter) that severely damaged or dying cells can express various kinds of genes including Oct4, as a epiphenomenon of dying process, in a random manner. This hypothesis well explains why gene expression level of Oct4 is relatively lower (2.5 or 8 folds) than usual positive control, such as iPS cells or ES cells. Ironically, this phenomenon can also explain well what Dr. Obokata and coauthors saw in their lab as STAP phenomenon -- actually it could be merely a epiphenomenon of dying process.

    By the way, I'm Buddhist and I can understand the meaning of “迴光返照. Buddhists believe we'll experience "death throes" in dying process, but the current result may suggest we'll get miraculous power at the time. :P)

    I'm afraid this report can be mistaken as "STAP strikes back" in Japan so I have to wait until the situation allows to share this...

    Finally, I vote to 2. "April Fool Cells"!


    Many thanks,

    -TJO

    Modified Apr 15, 2014 by the commenter.

  • Henry Chang · National Institutes of Health
    It would be interesting to try these experiments with amphibian cells that are capable of regeneration after trauma.
  • Henry Chang · National Institutes of Health
    It would be interesting to try these experiments with amphibian cells that are capable of regeneration after trauma.
  • Lars E Gustafsson · Karolinska Institutet
    Have you considered that you might need cells from the exact strain of animals used in the original work, with the same microbiological background, including possible viral contamination? Compare with the history of diabetic-prone animals which turned out to carry an infection which made them prone to develop the condition.

    Negative results might just mean that all necessary conditions for observing the effect are not at hand.

    This question is in no way in support of or criticism of any of the work done so far, either in the original or follow-up efforts. 

    In appreciation of your oppenness and efforts,

    LG

Open Review

Join ResearchGate to comment on this review or create one of your own.

Join for free

Commenters (24)